There are over 250,000 kidney transplant recipients in the United States. Essentially all require ongoing immunosuppression to prevent T cell (Tc) and B cell (Bc) mediated graft rejection. Such medications alter Bc differentiation into PC and memory Bc, impeding Bc and antibody responses to viral pathogens and vaccines, including a risk of severe influenza infection. Understanding Bc differentiation and identifying differences between normal and immunosuppressed patient responses to influenza vaccination using quantitative approaches could significantly improve clinical vaccine composition and methods. This is a significant public health is- sue, as immunosuppressed transplant recipients are recommended to have annual influenza vaccination, even though the response rates may only be 50%. Furthermore, influenza infection carries a substantial morbidity and mortality for this population. Given that current experimental and clinical research methods only permit sampling of peripheral blood immune responses, our proposed approach of (1) novel statistical methods for analyzing vaccine-specific Bc in peripheral blood, combined with (2) in vitro and in vivo branching stochastic process modeling, offers a novel and quantitatively rigorous method of extracting more relevant information from such data. This project is designed to identify peripheral blood Bc subsets associated with vaccine responses healthy volunteers and immunosuppressed kidney transplant recipients. Our primary hypothesis is that circulating Bc populations after vaccination in renal transplant recipients who respond to vaccine and those who do not respond differ in respect to their phenotypes, and their activation, division, differentiation kinetic rates. The overall goal of this project is to test this hypothesis using noel statistical methods and branching process (BP) models of Bc differentiation combined with extensive in vitro and in vivo data collection. The specific goals are: (1) To develop novel statistical methods for the analysis of high dimensional flow cytometery data used to identify Bc populations, (2) To characterize the in vivo and in vitro differentiation of murine Bc after influenza vaccination using these novel statistical methods, (3) to develop branching stochastic process models of in vitro and in vivo murine Bc differentiation, and (4) To identify and study differences in Bc responses to influenza vaccination in healthy and immunosuppressed renal transplant recipients using the novel statistical methods and branching process models. We evaluate how well the statistical methods and the model predict the success or failure of influenza vaccination in healthy volunteers, immunosuppressed vaccine responders and non-responders.